In some cases, e.g. small brain tumor metastases, responses to single or multiple fraction high-dose radiation therapy (HD-RT) have been remarkable, suggesting that HD-RT tumor control is at least as effective as biologically equivalent doses of conventional fractionated radiation therapy (CF-RT), even in radioresistant tumors. Although the mechanism for its effectiveness is not well understood, HD-RT is becoming accepted practice for a variety of tumors, including brain tumors. Our recent preclinical study using MRI measures of short-term changes in tumor physiology after HD-RT in a small-animal model of cerebral tumor suggests a physiological response that includes vascular effects, but is multifactorial and temporally variable. Hypothesizing that these short-term changes may both explain the increased effectiveness of HD-RT, and serve as a predictor of long-term response, we propose to investigate the relationship between short-term physiological changes after HD-RT and long-term outcome as a result of that therapy. In counterpoint, we will also study physiological changes during and after CF-RT. Detailed poroelastic modeling is proposed to generate a map of local solid and fluid parameters (stress, flow) that will help explain short-term changes in physiology. Aim 1 studies short-term changes in measures of tumor physiology as predictors of response. Aim 2 describes the behavior of these same measures over the course of CF-RT. Our long-range goals are to develop noninvasive biomarkers of response that predict tumor control after HD-RT and CF-RT, and to describe physiological changes and related biomarkers that might be used to optimize the order and timing of RT and adjuvant chemotherapies.